Obuchowski-Rockette variance components for dataset

Obuchowski-Rockette variance components for dataset

UtilORVarComp(
  dataset,
  FOM,
  covEstMethod = "jackknife",
  FPFValue = 0.2,
  nBoots = 200,
  seed = NULL
)

Arguments

dataset

Factorial one-treatment or cross-modality two-treatment dataset

FOM

Figure of merit

covEstMethod

The covariance estimation method, "jackknife" (the default) or "bootstrap" or "DeLong" ("DeLong" is applicable only for FOM = "Wilcoxon").

FPFValue

Only needed for LROC data and FOM = "PCL" or "ALROC": the FPFValue at which to evaluate a partial curve based figure of merit. The default is 0.2.

nBoots

The number of bootstraps (default 200).Only needed for covEstMethod = "bootstrap".

seed

Only needed for the bootstrap covariance estimation method. The initial seed for the random number generator, the default is NULL, for random seed.

Value

A list containing the following data.frames:

  • foms: the figures of merit for different modality-reader combinations

  • TRanova: the OR modality-reader ANOVA table

  • VarCom: the OR variance-components Cov1, Cov2, Cov3, Var and correlations rho1, rho2 and rho3

  • IndividualTrt: the individual modality mean-squares, Var and Cov2 values

  • IndividualRdr: the individual reader mean-squares, Var and Cov1 values

Details

The variance components are identical to those obtained using St with method = "OR".

Examples

## use the default jackknife for covEstMethod
vc <- UtilORVarComp(dataset02, FOM = "Wilcoxon")

##UtilORVarComp(dataset02, FOM = "Wilcoxon", covEstMethod = "bootstrap", 
##nBoots = 2000, seed = 100)$VarCom 

##UtilORVarComp(dataset02, FOM = "Wilcoxon", covEstMethod = "DeLong")$VarCom

vc <- UtilORVarComp(datasetX, FOM = "wAFROC")